| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Institute of Food, Nutrition, and Human Health and
and
Institute of Veterinary, Animal, and Biomedical Sciences, Massey University, Palmerston North, New Zealand
2 Correspondence:
Private Bag 11 222 (phone: 64-6-350-4336; fax: 64-6-350-5657; E-mail:
R.Purchas{at}massey.ac.nz).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionImplicationsLiterature Cited |
|---|
Key Words: Age Differences • Beef cattle • Meat Quality • Tenderness
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionImplicationsLiterature Cited |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionImplicationsLiterature Cited |
|---|
Bulls and steers (n = 117) used in this experiment were spring-born in either 1996 (Lot 1; Hereford-Angus cross cattle from one farm, n = 60) or 1997 (Lot 2; Angus cattle from a second farm, n = 57). The two farms were chosen because they each had compact calving periods and because at least 80 male calves were available for animal selection. Selection took place at approximately 2 mo of age to provide groups of 60 that were moderately even in terms of weight. Birth weights and dates of birth were not recorded. Half of each lot was selected at random to be castrated at an age of about 2 mo, and cattle were transferred from the commercial farms of birth to a Massey University research unit immediately after weaning at an age of about 5 mo. After a 2-wk acclimation period, growth rates were measured over test periods of 100 and 108 d for Lots 1 and 2, respectively, so cattle could be sorted within lot and gender into the slower-growing third (the Slow group) and the faster-growing two-thirds. Within the faster-growing group, half were allotted randomly to the Fast group and the remainder comprised the Restricted group. Subsequent nutritional management was such that animals of the Fast group were slaughtered at an age of 16 to 18 mo. Animals in the Slow and Restricted groups were slaughtered an average of 269 d later at an age of 24 to 28 mo. Steers were slaughtered an average of 38 d later than the bulls in order to reduce differences in mean carcass weights. The cattle were grazed on mixed pastures comprised primarily of perennial ryegrass (Lolium perenne) and white clover (Trifolium repens) at all times, with some pasture hay being supplied during periods of pasture shortage during the winter. No hay was fed during the 3 mo prior to slaughter. The different growth patterns required by the experimental design were achieved by providing the treatment groups with varying amounts of similar pastures rather than by providing pastures that differed in quality.
Slaughter and Meat Sample Collection
On the day of slaughter for each group, cattle were removed from pasture at approximately 0800 and trucked 20 km to a meat plant where preslaughter handling, washing, percussion stunning, exsanguination, and dressing followed normal commercial procedures, including low-voltage electrical stimulation (30 s with 14.28 pulses of 7.5 ms•s-1 and a peak voltage of 90 V) immediately after exsanguination. The period off feed prior to reaching the meat plant was approximately 2 h and the holding time at the plant was either 4 or 28 h, with a random half within each lot by treatment by gender group being allocated to each holding-time class. There were five animals within each (lot x gender x treatment x holding time) subgroup, except for three cells that contained four animals each.
During the period prior to the carcasses entering the chiller, the number of erupted permanent incisor teeth was recorded, and for each side, the length from the distal end of the tarsal bones to the midpoint of the cranial edge of the first rib (carcass length), and the length from the caudal edge of the cut pelvic bone to the midpoint of the cranial edge of the first rib (body length) were measured. In addition, the kidney plus pelvic (channel) fat from each side was removed and weighed. The liver (without gall bladder) and heart (after removal from the pericardium) were also weighed. Additionally, a sample of 1,800 to 2,000 g of the longissimus muscle (LM) was excised between the 6th and 12th ribs of the right side immediately prior to chilling.
After overnight chilling of the carcass at 1 to 3°C (18 to 24 h postmortem), the cross-sectional area of the LM at a transverse cut between ribs 12 and 13 was traced, and the area subsequently measured using a digital planimeter (Placom KP-90N, Tokyo, Japan). The fat thickness over this muscle at the same site between ribs 12 and 13 and at two-thirds of the muscle width from its medial edge was measured. The left femur bone was collected during fabrication, and after having muscle and fat remnants removed, its weight and maximal length were recorded as an indication of skeletal development.
Laboratory Methods
Muscle samples collected within 90 min postmortem were held at 12 to 15°C for 24 h to avoid any cold-shortening. These were then aged at 2 to 4°C for a further 6 d before being divided into an 800-g anterior portion for sensory evaluation, which was vacuum-packed prior to freezing at -30°C, and a posterior portion (~1,000 g), which was packaged and frozen at -20°C until objective measures of tenderness and other meat quality characteristics could be made. After thawing the latter sample overnight at 2 to 3°C, two 25-mm thick steaks were cut for assessment of Warner-Bratzler shear force and Meat Industry Research Institute of New Zealand (MIRINZ) force score after cooking at 70°C for 90 min (Purchas and Aungsupakorn, 1993). A third 25-mm thick steak was cooked at 60°C for 60 min for assessment of compression parameters. At the time of preparing these steaks, uncooked samples were set aside for the evaluation of meat color, muscle pH, myofibrillar fragmentation index, sarcomere length, muscle fiber diameter, and intramuscular fat.
For the Warner-Bratzler and MIRINZ tenderometer measurements, six 13-mm2 cores were prepared from each steak in such a way as to ensure that shears were made across the fibers. Measurements with a Warner-Bratzler device (crosshead speed of 230 mm•min-1; G-R Eletric Mfg. Co., Manhattan, KS) fitted with a square blade and a 30-kg load cell as described by Purchas and Aungsupakorn (1993), were made at about one-third the distance along each core. Parameters recorded were total work done (WB-WD), initial yield (WB-IY), and peak force (WB-PF) (Purchas and Aungsupakorn, 1993). The remaining two-thirds of each core was then trimmed to a 10 x 10 mm cross section and assessed with the MIRINZ tenderometer (Smith-Biolab, Ltd., Auckland, NZ) for a force score (Macfarlane and Marer, 1966).
An Instron compression device was used to measure hardness (Harris, 1976) as the force required to drive a 10 x 10 mm square-faced flat-ended plunger (at a crosshead speed of 100 mm•min-1) 8 mm into a sample of meat cooked at 60°C for 60 min with a cross section across the fibers of 10 x 10 mm, as described by Peachey et al. (2002).
Muscle pH was measured in an homogenate of 2 g of LM in 10 mL of 5 mM iodoacetate and 150 mM KCl (Bendall, 1973). Sarcomere length was measured by laser diffraction (Bouton et al., 1973). Meat color was assessed with a Minolta Chromameter (CR-200, 8 mm measured area diameter, standard illuminant C, white tile calibration; Ramsey, NJ) following exposure of a cut surface to air for at least 30 min. Measurements of intramuscular fat by Soxhlet extraction and expressed juice by the filter paper press method were as described by Purchas and Aungsupakorn (1993). Measurements of myofibrillar fragmentation index (MFI%) by a filtration method based on that of Johnson et al. (1990) were carried out as outlined by Purchas et al. (1997). The MFI% procedure included a drying step so that values ranged from 78 (when no fragments passed through the filter) to 100 (when all fragments passed through). Muscle fiber diameter was measured on the same slides that were prepared for sarcomere length assessment. The diameter of 13 fibers that extended as single fibers from the edge of the preparation on the slide were measured for each sample using an eye-piece micrometer fitted to a microscope at 100x.
Sensory evaluation was by a trained panel of 11 people, each of whom evaluated three steaks from each animal over two series of 15 sittings with 12 samples being evaluated by each panelist at each sitting. Details of sensory panel procedures, including the training protocol, were reported by Peachey et al. (2002). Characteristics evaluated on 100-mm unstructured line scales included initial juiciness, hardness, cohesiveness, toughness, chewiness, and overall juiciness as described by Peachey et al. (2002).
Statistical Analysis
Statistical analyses were performed using the GLM procedure of SAS (SAS Inst., Inc., Cary, NC). The main model was a 3 x 2 x 2 x 2 factorial with three treatments (Fast, Restricted, and Slow), two lots (yr 1 and yr 2), two genders (bulls and steers), and two holding times at the meat plant (4 or 28 h). Covariates fitted before the main effects in the models included carcass weight for composition characteristics, and ultimate pH for meat quality characteristics. All two-way interactions between the main effects and between the main effects and covariates (when present) were evaluated, but were not retained in the model if they were not significant (P < 0.05).
Data from the sensory panels were initially analyzed using the raw data to derive least squares means for each animal, and then these means were used in subsequent analyses to evaluate the significance of the main effects and the relationships between sensory characteristics and other measurements (Peachey et al., 2002). The design used to determine which samples to present to panelists at a particular sitting, and the sequence in which the samples were presented, permitted the effects of different panelists and sequences to be adjusted for in the first analysis. A session effect could not be taken out because all panelists received samples from the same 12 animals at the same sitting. Animal and session effects were not confounded, however, because each panelist evaluated three samples from each animal in different sessions, and different combinations of animals were used at each sitting.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionImplicationsLiterature Cited |
|---|
On-Farm Performance
Growth rates for bulls and steers within the three treatment groups are shown separately for Lot 1 and Lot 2 in Table 1
over four periods. Growth rates during the initial "test period" were used to allocate animals to treatment groups as described above. During the differential-feeding period up to the time when the first animals from the Fast group were slaughtered, the Fast group was given access to more pasture than the other two groups so that they could be slaughtered before their second winter. The final two periods, with only the Restricted and Slow groups present (Table 1), were up to the time when the first of the animals in those groups were slaughtered. This period included an initial period of slow growth over the winter followed by a period of faster growth during the spring leading up to slaughter.
|
) due to the difference between bulls and steers being greater within the Fast group than within the Restricted and Slow groups.
Withers heights did not differ significantly between the treatment groups when values were adjusted to the same live weight, but weight-adjusted withers heights were greater for steers than bulls (P < 0.001) at the three older ages when they were measured (Figure 1). These results are consistent with those of Brannang (1971), who showed that the long bones of bulls were shorter than those of steers, possibly due to the action of testosterone on the epiphyseal growth plate.
|
Despite the fact that steers were slaughtered an average of 38 d later than the bulls, their mean carcass weight was 18.3 kg less (Table 2). There were no (P > 0.10) carcass weight differences between the three growth-pattern groups. Weight-adjusted dressing percentages were slightly higher (P < 0.10) for bulls than steers, as has been reported in some other studies (Field, 1971; Purchas and Grant, 1995), and were higher (P < 0.05) for the Fast group than for the Slow or Restricted groups, probably because of higher (P < 0.05) levels of fatness (Table 3).
|
|
). Such effects have been shown previously (Purchas and Grant, 1995) and reflect the fact that cattle tend to continue to grow in frame size even when overall body growth has been restricted by lower levels of nutrition. Differences between groups in the weights of heart and liver were small, but hearts were heavier (P < 0.05) for bulls than steers, and the Fast-group cattle had lighter livers than animals in the Slow and Restricted groups (P < 0.001). Teeth eruption patterns at the time of slaughter (results not shown) revealed no clear differences between the bulls and steers. All animals in the Fast group and 23.1% of those in the Restricted and Slow groups had no permanent incisors erupted at the time of slaughter, which indicates that the different tenderness of beef from cattle in these groups could not have been predicted satisfactorily on the basis of teeth eruption patterns if the age of the animals was not known.
Results relating to the fatness and muscling of the cattle after adjustment to a common carcass weight are given in Table 3. As expected, steers were fatter (P < 0.001) than bulls at the same carcass weight for all measures of fatness, and they had smaller (P < 0.01) LM areas, but there was no castration effect on femur bone weight. The Fast group of cattle was fatter than the other two groups in terms of kidney and pelvic fat weight (P < 0.05), fat thickness (P < 0.05), and intramuscular fat level (P < 0.05) (Table 3), and had significantly lighter (P < 0.05) femur bones at the same carcass weight. Longissimus muscle areas did not differ (P > 0.10) between the growth-pattern groups. There are numerous reports of cattle being less fat at a common carcass weight when they have undergone a period of restricted feeding prior to being finished; although in some studies, it is difficult to separate the effects of growth pattern from the effects of different diets. Carstens et al. (1991) found that Angus-Hereford cross steers that had been through a period of restricted growth prior to finishing were less fat (24 vs 32% empty body fat) than a nonrestricted group of the same final weight. Purchas and Grant (1995) demonstrated that cattle from the same initial population that were slaughtered at 20 mo were fatter than those slaughtered at 28 mo after adjustments were made for differences in carcass weight. Steen and Kilpatrick (1995) reported that cattle finished on a ration of grass silage and concentrates offered ad libitum grew faster and had greater fat depths at the same carcass weight (+35% for bulls and +14% for steers) than cattle fed at a rate of 80% ad libitum. Short et al. (1999) found that steers of several crosses placed on a high-concentrate finishing diet at 6 mo of age had fat depths and marbling scores approximately double those for cattle placed on-feed at 12 mo of age, when comparisons were made at a carcass weight of about 250 kg.
Sensory and Objective Measures of Beef Tenderness
Results using both subjective and objective measurements were very consistent in showing that beef from bulls was tougher (P < 0.01) than that from steers (Tables 4 and 5 ). Sensory evaluation of the four aspects of toughness assessed (hardness, cohesiveness, toughness, and chewiness) were higher (P < 0.001) for the bull beef (Table 4). In addition, bull beef was assessed as being less juicy (P < 0.05), particularly in terms of initial juiciness, which may be attributable in part to the fact that intramuscular fat levels were only about 30% of those for steer beef. Juiciness was included in the sensory analysis because it has been shown to be an integral component of the sensory perception of meat texture (Mathoniere et al., 2000).
|
|
). Multiple objective measures were employed because different components of tenderness were measured by the different devices. For example, the difference between WB-PF and WB-IY has been shown to be indicative of the connective tissue contribution to tenderness (Beilken et al., 1986), and the Instron compression measures should have been indicative of connective tissue characteristics because of the lower cooking temperature and a shorter cooking time, as well as the nature of the measurement (Eikelenboom et al., 1998). The fact that the difference for the compression measures was at least as great as for other measures of tenderness, indicates that collagen differences may have been involved in toughness differences between beef from bulls and steers, as has been suggested in other studies (Dikeman et al., 1986; Bosselmann et al., 1995).
Ultimate pH as a linear and quadratic term was included as a covariate for all the characteristics shown in Tables 4 and 5 because it had a significant effect on most of the characteristics. The nature of the relationship with measures of tenderness was similar to that reported elsewhere (e.g., Purchas and Aungsupakorn, 1993) with maximal toughness at a pH of around 6.0. When ultimate pH was left out of the model, the coefficients of determination (R2) were significantly reduced, but the nature and significance of differences between the groups remained essentially the same.
Differences between the three growth-pattern groups for measures of beef tenderness were smaller than the differences between bulls and steers, but generally, beef from the Fast group was more tender than that from the Slow and Restricted groups based on sensory scores (Table 4) and objective measures (Table 5). In some cases (WB-IY, WB-WD, and MIRINZ force score), the Restricted group was not significantly different from either of the other groups. Unlike the bull/steer comparison, growth pattern did not affect the Instron compression parameters (Table 5) or measures of sensory juiciness (Table 4). Interactions between growth pattern and castration status were not significant. Means within the subgroups for two measures of tenderness (Figure 2) illustrate how the Fast group bulls produced beef that was not significantly tougher than beef from steers of the Slow or Restricted groups.
|
Other Quality-Related Characteristics
Beef from Bulls and Steers.
Ultimate pH values were higher (P < 0.001) for bulls than steers (Table 6), but the difference of 0.19 pH units was smaller than for some previous studies. Purchas and Grant (1995), for example, reported mean values of 5.64 and 6.07 for Friesian-cross steers and bulls, respectively. The higher (P < 0.05) ultimate pH for the Fast group relative to the Slow and Restricted groups (Table 6) was largely attributable to a particularly large difference between bulls and steers in the Fast group (5.88 for bulls vs 5.50 for steers), giving rise to a significant bull/steer x growth pattern interaction for ultimate pH (P < 0.01). All characteristics in Table 6, except MFI%, were significantly affected by ultimate pH and were, therefore, adjusted to a common ultimate pH. This gave models with a better fit (higher R2 values), and including adjustments for ultimate pH had minimal effects on the size or significance of differences between castration groups or growth-pattern groups for measures of tenderness or most other quality characteristics. This contrasts with the results of Purchas and Aungsupakorn (1990), where differences between bulls and steers for several quality characteristics changed when adjustments for ultimate pH were made; but in that study, the group difference in mean pH was much larger.
|
Beef from bulls and steers did not differ with respect to pH-adjusted lightness (L*), but beef from steers was redder (P < 0.001) and more yellow (P < 0.001) than beef from bulls. When values were not adjusted to constant pH, L* values were also lower for bulls (35.0 vs 36.8, P < 0.01). The lower MFI% values were consistent with studies showing lower proteolytic enzyme activity in muscle from bulls (Morgan et al., 1993; Thompson et al., 1996). The lower water-holding capacity of bull beef after adjustment for pH differences may partially explain the lower levels of tenderness for beef from bulls relative to that from steers. Lower pH-adjusted levels of cooking loss have been reported previously for beef from bulls (Dikeman et al., 1986; Purchas et al., 1997), whereas differences between beef from bulls and steers for cooking losses and expressed juice values were not significant when data were not adjusted for pH (data not shown). Similar results have been reported previously by Purchas and Aungsupakorn (1993) and Purchas and Grant (1995).
A clear explanation of why the beef from bulls was less tender than that from steers in this study was not apparent, but the lower proteolytic activity, slightly higher ultimate pH, lower levels of intramuscular fat, higher cooking losses, and possibly a greater contribution of connective tissue components may all have contributed. Comparisons of the tenderness of beef from bulls and steers have been reported many times with the results being highly variable, as noted in reviews by Field (1971), Seideman et al. (1982), and Purchas (1991). The tenderness difference reported here was larger than that in most previous studies, and although it is supported by some other reports (Crouse et al., 1983; Gregory et al., 1983; Dikeman et al., 1986), it also needs to be considered alongside many other studies where significant differences have not been detected (Woodhams and Trower, 1965; Dransfield et al., 1984; Purchas and Grant, 1995). Clear explanations for the basis of these variable results would be of value when seeking ways to produce beef of consistently high quality from bulls.
Growth-Pattern Effects.
Color differences between the three growth-pattern groups were small (Table 6), with no difference in lightness, but with the Fast group producing beef that was slightly redder and more yellow (P < 0.05) after an adjustment for differences in ultimate pH. Beef from the Fast group also had a significantly higher MFI% (P < 0.05), suggesting that proteolytic activity was greater, which could go some way to explaining the higher tenderness scores. There were no differences (P > 0.10) in sarcomere length or fiber diameter, and measures of water-holding capacity were similar (P > 0.10), except that cooking loss at 70°C was lower (P < 0.05) for the Fast group than the Slow and Restricted groups. Overall, there was no single characteristic that stood out as being responsible for the greater tenderness of beef from the Fast group for both bulls and steers (Figure 2), and it seems likely that the difference may have been the net effect of a number of factors, including higher MFI% and intramuscular fat level, lower cooking loss, and possibly, differences in the nature of the collagen that were not measured directly in this study. The fact that there was little difference in the tenderness of beef between the Restricted and Slow groups indicates that their lower tenderness relative to the Fast group was primarily due to their greater age and/or to the different growth paths.
Clear reductions in beef tenderness have been reported when wide ranges in age have been considered (Hiner and Hankins, 1950 [2.5 to 67 mo of age]; Tuma et al., 1962 [18 mo to 90 mo of age]), and especially when muscles with a high connective tissue content are considered (Shorthose and Harris, 1990), apparently largely due to increases in collagen cross-linking (Goll et al., 1964; Nishimura et al., 1999). Over shorter age ranges, such as those involved in the current trial (8 to 10 mo), there have been reports of increases, decreases, and no change in measures of tenderness. Thus, Jacobson and Fenton (1956) reported that tenderness of beef from Holstein heifers decreased as age increased from 11 to 20 mo, and Field et al. (1966) showed that beef was tougher for bulls aged 600 to 699 d relative to those aged 300 to 399 d, but was more tender for steers and heifers of similar ages. No significant difference in tenderness was noted between cattle aged 12 or 24 mo by Dikeman et al. (1986) or by Gullet et al. (1996), but Short et al. (1999) showed that tenderness improved with increased time on feed (and therefore increased age), particularly for cattle that entered the feedlot at 12 mo of age. Vestergaard et al. (2000) reported that beef from bulls raised in an extensive system to 14 mo was less tender than beef from bulls raised intensively to the same weight by 11 mo. Muir et al. (2001) reported that pasture-fed steers that had undergone a period of slow growth over winter yielded beef that was less tender than that from control cattle that had grown faster and were therefore younger at slaughter. Oddy et al. (2001), in reviewing the effect of patterns of growth, concluded that there was good evidence that growth patterns could influence beef tenderness, but that the exact nature of the effects could not be accurately predicted. They did note, however, that high rates of growth leading up to slaughter, as was the case for all groups in the current trial (Table 1), were generally associated with reduced toughness. Sinclair et al. (2001), on the other hand, concluded that there was limited scope for improving beef eating quality by increasing growth rate through higher levels of nutrition.
Holding-Time Effects
Cattle held at the meat plant for an extra 24 h had lower carcass weights (P < 0.01) (Table 7), which was probably due to a decrease in gut contents (Kirton et al., 1972). This was reflected in a lower (P < 0.001) dressing percentage based on live weight at the time of slaughter for the 4-h hold group. Dressing percentage based on live weight prior to trucking, however, was higher (P < 0.001) for the 4-h hold group, indicating that some of the weight lost over the extra 24 h of holding time was carcass weight. For a 500-kg animal, this decrease in dressing-out percentage would represent a decrease of 6.9 kg in carcass weight. The suggestion by these results that significant carcass weight losses may occur after fasting times of 30 h or less, is in agreement with some other studies (Raikes and Tilley, 1975; Jones et al., 1990; Purchas, 1992), but not with others (Carr et al., 1971; Kirton et al., 1972), where longer periods off feed have been needed before carcass weight losses have been detected.
|
None of the pH-adjusted measures of beef tenderness, juiciness, or color differed between the two holding-time groups. Three of these measurements are given as examples in Table 7. Similar results were obtained for unadjusted values (data not shown). Beef from cattle held for the extra 24 h had lower (P < 0.001) expressed juice levels and lower (P < 0.05) cooking losses at 70°C when adjustments were made for differences in ultimate pH. The differences remained when adjustments for pH were not made (data not shown). These results are consistent with a loss of muscle water during the longer holding period, as was suggested by the studies of Jones et al. (1988).
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
Received for publication April 15, 2002. Accepted for publication August 9, 2002.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. Mach, A. Bach, and M. Devant Effects of crude glycerin supplementation on performance and meat quality of Holstein bulls fed high-concentrate diets J Anim Sci, February 1, 2009; 87(2): 632 - 638. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
